This is an implementation of the "Think Tool" described in Anthropic's engineering blog post as an MCP server. The Think Tool is a simple but effective prompt engineering technique that helps Claude break down complex problems and enhance its reasoning capabilities.
How it Works
The Think Tool is incredibly simple - it provides a no-op tool that does nothing except echo back the input. The magic is in how it allows Claude to:
- Take a step back and think through complex problems
- Break down reasoning into discrete steps
- Organize thoughts more systematically
- Cache intermediate results during complex calculations
- Show its work when solving problems
As described by Anthropic, this is a "prompt engineering trick" where they use the standard tool calling mechanism to define a tool called "think" that "does nothing at all" - there is no implementation - it simply allows the model to use its existing training about when to use tools to stop and dump additional thoughts into the context.
Implementation Details
The MCP server exposes a single tool:
think - Takes a thought as input and returns it
Tool definition:
{
"name": "think",
"description": "Use the tool to think about something. It will not obtain new information or change the database, but just append the thought to the log. Use it when complex reasoning or some cache memory is needed.",
"input_schema": {
"type": "object",
"properties": {
"thought": {
"type": "string",
"description": "A thought to think about."
}
},
"required": ["thought"]
}
}
Usage
Setup
- Clone this repository
- Run
npm install to install dependencies - Run
npm run build to compile TypeScript - Run
npm start to start the MCP server
Connecting to Claude Desktop
Add this server to your Claude Desktop configuration file:
{
"mcpServers": {
"think-tool": {
"command": "node",
"args": ["/path/to/mcp-think-tool/build/index.js"]
}
}
}
Add this prompt to teach the LLM how to use the think tool:
## Using the think tool
Before taking any action or responding to the user after receiving tool results, use the think tool as a scratchpad to:
- List the specific rules that apply to the current request
- Check if all required information is collected
- Verify that the planned action complies with all policies
- Iterate over tool results for correctness
Here are some examples of what to iterate over inside the think tool:
<think_tool_example_1>
User wants to cancel flight ABC123
- Need to verify: user ID, reservation ID, reason
- Check cancellation rules:
* Is it within 24h of booking?
* If not, check ticket class and insurance
- Verify no segments flown or are in the past
- Plan: collect missing info, verify rules, get confirmation
</think_tool_example_1>
<think_tool_example_2>
User wants to book 3 tickets to NYC with 2 checked bags each
- Need user ID to check:
* Membership tier for baggage allowance
* Which payments methods exist in profile
- Baggage calculation:
* Economy class × 3 passengers
* If regular member: 1 free bag each → 3 extra bags = $150
* If silver member: 2 free bags each → 0 extra bags = $0
* If gold member: 3 free bags each → 0 extra bags = $0
- Payment rules to verify:
* Max 1 travel certificate, 1 credit card, 3 gift cards
* All payment methods must be in profile
* Travel certificate remainder goes to waste
- Plan:
1. Get user ID
2. Verify membership level for bag fees
3. Check which payment methods in profile and if their combination is allowed
4. Calculate total: ticket price + any bag fees
5. Get explicit confirmation for booking
</think_tool_example_2>
Replace /path/to/mcp-think-tool with the actual path to this repository.
Examples
Claude might use the Think Tool to work through a problem like:
Solving 235 × 47:
Think: First I'll break this down. I need to multiply 235 by 47.
Think: I'll start by calculating 235 × 40 = 9,400
Think: Then I'll calculate 235 × 7 = 1,645
Think: Now I add them together: 9,400 + 1,645 = 11,045
Therefore, 235 × 47 = 11,045
Or for a more complex reasoning task:
Think: I need to analyze the given problem carefully. The question asks about the impact of reducing carbon emissions by 15% over 5 years.
Think: First, I should establish the baseline emissions. The document mentions current annual emissions of 50 million metric tons.
Think: 15% reduction over 5 years means approximately 3% reduction per year, assuming linear decrease.
Think: After 5 years, annual emissions would be 50 × (1 - 0.15) = 50 × 0.85 = 42.5 million metric tons.
Think: The total reduction over 5 years would be the sum of the reductions each year...
Development
src/index.ts - MCP server implementation
License
ISC